Effect of alternative fuel properties on NO x reduction
M. Sc. Axelsen, Ernst Petter
Scancem International ANS1, Technical Support Setreveien, NO-3950 Brevik, Norway
Dr. ing. Tokheim, Lars-André NORCEM AS1
Setreveien, NO-3950 Brevik, Norway
Professor, Dr. ing. Bjerketvedt, Dag Telemark University College, Faculty of Technology
Kjolnes Ring 56, NO-3914 Porsgrunn, Norway
Introduction
Today we see a substantial increase in the use of alternative fuels in the cement industry. The prospect of reduction in fuel costs and the environmental benefits of waste to energy conversion are the driving forces. For several years Norcem have steadily increased their use of alternative fuels such as refuse derived fuel (RDF), liquid hazardous waste (LHW), solid hazardous waste (SHW), animal meal (AM) and waste oil (WO). Alternative fuels behave differently compared to e.g. coal during combustion, which is a factor that may influence the clinker quality, kiln operation, energy consumption and emissions. Scancem International ANS and Telemark University College has therefore initiated a research program to provide a better basis for characterization and understanding of alternative fuels in cement kilns. The study focus on the effect of alternative fuel properties on NOx reduction by advanced reburning and reburning in conjunction with CO emission.
A laboratory circulating fluidized bed combustor (CFBC) is applied, in addition to full scale experiments at kiln 6 Norcem Brevik. Furthermore, computational fluid dynamic (CFD) simulations with the aid of Fluent ® on the flow regime of the precalciner at kiln 6 Norcem Brevik is performed.
This paper will summarize the findings including results from full-scale experiments with advanced reburning, animal meal and solid hazardous waste as a reburning fuel and with comparable tests in the CFBC reactor.
Full scale experiments
The precalciner kiln 6 in Brevik is a so called "in-line calciner" (ILC), where the off-gas from the kiln is led through the precalciner. A sketch of the precalciner kiln is shown in Figure 1. Table 1, 2 and 3 show the characteristics of the kiln, the fuels and the emissions of kiln 6, respectively.
Figure 1: Precalciner kiln 6, Brevik
1 Part of Heidelberg Cement
Table 1: Characteristics of kiln 6,Brevik.
Kiln length 68 m
Outer kiln diameter 4.4 m
Nominal production capacity 3300 tons clinker/day Typical specific fuel consumption 3300 kJ/kg clinker Approximate static pressure loss in the preheater 7000 Pa Specific cooler load at nominal production capacity 48 tons clinker/m²day Table 2: Fuel characteristics of kiln 6,Brevik.
Parameters Unit Coal LHW SHW RDF AM WO
Moisture % (As rec.) 2.80 45.70 10.70 25.00 4.50 0.00
Volatiles % 31.60 100.00 46.30 53.0 59.70 100.00
C.fix % 45.5 0.00 7.70 10.50 7.50 0.00
Ash % 20.1 0.00 35.30 11.50 28.30 0.00
LHV MJ/kg (as rec.) 24.74 17.88 15.75 18.0 15.80 41.60
O/N Molar ratio 2.30 11.89 36.65 34.60 1.57 -
Table 3: Emission characteristics of kiln 6,Brevik.
Emission component SFT limit value2 mg/Nm3@ 11 % O2
2000 mg/Nm3@ 11 % O2
Dust 40 5
Total organic carbon (TOC) 10 6.5
Chlorine compound calculated
as HCl 10 6.6
Fluorine compound calculated
as HF 1 0.1
SOx calculated as SO2 300 170
NOx calculated as NO2 707
Sum Cd + Tl 0.1 0.081
Mercury 0.1 0.002
Sum of other metals3 1.0 1.913
Dioxines 0.14 0.0255
Execution of full scale experiments
The experiments in the precalciner have been executed in two campaigns in order to obtain reburning and advanced reburning, according to Table 4, 5 and 6. Henceforth, refereed to as experiment 1 to 7 for the experiments in February and 8 to 17 for the experiments in April.
Advanced reburning is obtained with urea pellets fed into the precalciner at the kiln gas side with the alternative fuel (SHW). The proportioning of urea pellets are executed with a screw feeder into the alternative fuel feed stream.
The reburning fuels are solid hazardous waste and animal meal.
2 The Norwegian State Pollution Control Agency (SFT)
3 Sb, As, Pb, Cr, Co, Cu, Mn, Ni, V, Sn
4 I-TE ng/Nm3
5 I-TE ng/Nm3
Table 4: Average fuel- and solid feed into the precalciner kiln 6 during experiments in February 2001.
No
feed SHW SHW + 38 kg/h urea
SHW + 153 kg/h
urea
SHW + 269 kg/h
urea
SHW + 386 kg/h
urea
AM Parameters
#1 #2 #3 #4 #5 #6 #7
CoalPrimary Burner t/h 4.1 3.6 3.8 3.8 3.6 3.5 4.2
CoalPrecalciner t/h 9.9 9.0 9.7 8.8 9.2 9.0 8.9
LHW t/h 1.5 1.7 1.8 1.8 1.8 1.8 1.5
SHW t/h 0.0 3.0 3.0 3.0 3.0 3.0 0.0
RDF t/h 2.9 2.5 2.0 2.3 2.0 2.0 3.1
AMPrecalciner t/h 0.0 0.0 0.0 0.0 0.0 0.0 2.6
AMPrimary Burner t/h 0.0 0.0 0.0 0.0 0.0 0.0 0.0
WO t/h 2.5 2.6 2.2 2.2 2.5 2.7 2.5
Raw Meal t/h 238 238 237 224 237 237 239
The reference condition is no feed of alternative fuel into the kiln gas side of the precalciner, with reference to experiment 1 and 8. Experiment 3, 4, 5 and 6 as well as experiment 10, 11, and 12 show increasing feed of urea to obtain advanced reburning (AR) rich with a 30 minutes interval.
Table 5: Average fuel- and solid feed into the precalciner kiln 6 during experiments in April 2001.
No feed
SHW SHW + 153 kg/h urea
SHW + 386 kg/h urea
SHW + 770 kg/h urea
AM Parameters
# 8 # 9 # 10 # 11 # 12 # 13
CoalPrimary Burner t/h 6 8 8 8 8 8
CoalPrecalciner t/h 8 5 6 5 6 6
LHW t/h 2 0 0 0 0 0
SHW t/h 0.0 4 4 4 4 0.0
RDF t/h 4 6 5 6 6 6
AMPrecalciner t/h 2.0 0.0 0.0 0.0 0.0 3
AMPrimary Burner t/h 0.0 0.0 0.0 0.0 0.0 0.0
WO t/h 0 0 0.0 0.0 0 2
Raw Meal t/h 220 218 220 220 220 226
It is accomplished two longer trials with advanced reburning rich in experiment 16 and 17 at two and one hour interval, respectively. It is essential to investigate the effect of reburning with SHW in comparison to the advanced reburning, shown in experiment 2 and 9. Utilization of animal meal in the precalciner is shown in experiment 7 and 13. In experiment 16 and 17 the effect of animal meal in the main burner is shown with and without urea in combination with SHW in the precalciner.
Table 6: Average fuel- and solid feed into the precalciner kiln 6 during experiments in April 2001.
SHW SHW + 386
kg/h urea SHW + 153
kg/h urea SHW + 386 kg/h urea Parameters
# 14 # 15 # 16 # 17
CoalPrimary Burner t/h 7 6 7 7
CoalPrecalciner t/h 5 4 5 4
LHW t/h 1 1 1 2
SHW t/h 4 4 4 4
RDF t/h 6 6 5 6
CoalPrecalciner t/h 0.0 0.0 0.0 0.0
CoalPrimary Burner t/h 2 2 0.0 0.0
WO t/h 1 0 0 0
Raw Meal t/h 213 220 220 220
Results and discussion of the full scale experiments
The principal trend from the campaign in February (experiment 1 to 7) and April (experiment 8 to 13) can be seen in Figure 2 and 3. Where, the principal development shows decreasing NOx emissions and increasing CO emissions.
The decreasing NOx emission can be explained with:
• reburning with SHW (experiment 2 and 9),
• increasing supply of urea together with SHW (advanced reburning rich) in experiment 3 to 6 and 10 to 12, and
• utilization of animal meal in experiment 7 and 13.
Additional supply of animal meal into the main burner during reburning and advanced reburning showed additional NOx reduction in experiment 16 and 17.
Figure 2: NOx emitted from kiln 6 during experiment 1 to 17.
Figure 3:CO emitted from kiln 6 during experiment 1 to 17.
The full-scale experiments showed that:
• advanced reburning and combustion of animal meal in the precalciner is well suited for NOx
reduction in the precalciner kiln 6 at Norcem Brevik, and
• combustion of animal meal in the main burner gives additional overall NOx reduction and lower NOx emissions at the kiln inlet although it result in a significant increase of fuel- nitrogen supply.
An increasing development of the molar ratio (urea/NOkiln inlet) lower the NOx emissions and that an optimum NOx reduction is at a ratio between 2.6 to 3, which is lower compared to previous experiments /1/.
The assumption is that combustion of animal meal follows a similar reduction route as advanced reburning or selective non-catalytic reduction because of a low HCN/NH4 ratio.
Hämäläinen (1995) showed that the/NH3 ratio in pyrolysis gases decreases with the fuel O/N ratio /2/.
It can be seen from Table 2 that the O/N ratio is low for animal meal and coal in comparison with solid hazardous waste (SHW) and refuse derived fuel (RDF), which signify a smaller HCN/NH3 ratio.
Experiments showed that up to 86 % of the nitrogen in animal meal is released during pyrolysis, compared to only 4 % in the coal.
Experiments executed by Zevenhoven et al. (2000) showed that smaller HCN/NH3 ratio favour NOx
reduction /3/. Comparing experiments made by Desroches-Ducarne et al. (1998) to experiment 7 and 13, shows that adding a fuel with low HCN/NH3 ratio in combination with RDF lowers the NOx
emissions /4/. Hence, NOx reduction with animal meal is most likely to originate from NH3 released during volatilisation.
The CO and TOC emissions follow the same pattern, as it is a result of incomplete combustion.
The increased CO emissions are probably due to the increasing amount of NH3 in the process during supply of urea and animal meal in experiment 1 to 17.
Equation 1 and 2 are the primary reduction reactions for NH3 and CO, respectively. It can be seen from Figure 4 that the primary abstraction reaction constant for NH3 is about ten times faster than the CO oxidation. Hence, the competition for the OH radical between Equation 1 and 2 are probably the reason for the increased CO emissions.
The increased TOC emissions can probably be due to the formation of e.g. NH4Cl.
(1)
O H NH OH
NH
3+ →
2+
2 (2)The ammonia slip is about 34 % of the NOx reduction during combustion with animal meal.
Furthermore, the ammonia slip is a minor part, 0.8 %, of the NOx reduction when animal meal is supplied into the main burner.
Figure 4: Comparison between the reaction rate constants for reaction: NH3+OH->NH2+H2O and CO+OH->CO2+H /5/.
H CO OH
CO + →
2+
Laboratory experiments
Laboratory experiments have been accomplished at Telemark University College with the use of a laboratory CFBC reactor. The main purpose is to investigate different fuels characteristics during:
• reburning,
• advanced reburning and
• combustion without circulating mass.
The laboratory CFBC reactor are designed and operated in order to obtain similar conditions as in a precalciner /5/. The CFBC reactor has a physical design of about 6 meter in height and 100 mm diameter.
A literature review showed that there is a similarity between the CFBC reactor and the precalciner regarding the phenomena of:
• gas- and solid mixing,
• core-annulus flow,
• micro flow structure and
• effects of particles on turbulence.
There is a similarity between the CFBC reactor and the precalciner at Norcem Brevik regarding the dimensionless groups applied, which roughly obtains the same:
• fluidization regime,
• riser solids hold-up by volume fraction and
• macroscopic movements of solid.
The largest difference between the CFBC reactor and the precalciner is the turbulence. The
consequence of not achieving similar turbulence will probably affect the particle retention time and the diffusion of oxidizer to the fuel particle.
accumulation / sampling CFB
reactor
PT D
A PI
Cyclone Ceramical filter
NOTES:
1. Perforated gas distribution plate 2. Air and fuel gas plenum chamber 3. Explosion hatch
4. Temperature transmitter (10 sensors in one unit;
from T1..T10). Center mont. inside CFB reactor 5. BFB reactor
6. Diaphragm 7. Peep hole
A TT TI D
TT A
D TI
Note 1 Note 3
Note 4
Note 5
Note 2
Secondary air
PT D A PI
Primary fuel (Propan) Primary combustian air Alternative fuel Alternative inlet of
circulating mass
Circulating mass
Analyser HC O2 CO CO2 N2O NO/NOx SO2
PI A
D PT
Exhaust Particle filter
+ tertiary air
+ tertiary air
Gas inlet Note 6
Note 7
Figure 5: Sketch of the CFBC reactor /7/.
Execution of laboratory CFBC experiments
In the experiments with reburning and advanced reburning it was of interest to investigate alternative fuels behavior in a precalciner environment i.e. with circulating mass6 and supply of up to 2000 mg/Nm³ of NO into the CFBC reactor. In the experiments without circulating mass fuel specific differences were expected.
It was decided to vary the supply of NO in four steps for the experiments with advanced reburning and combustion without circulating mass to investigate the impact of increasing supply of NO.
The reburning experiment was executed with virginal and blends of fuels, which was the main object with this set of experiments. The reburning experiment was executed with only one step of NO supplied (about 1930 mg/Nm³).
Table 7 show the principal operating conditions during the experiments including the levels of SO3, Cl and the calcination degree.
It was chosen to run the reburning and advanced reburning experiments with a particle size less than 1 mm. The experiment without circulating mass was tested with two sets of particle sizes, namely less than 1 and 2 mm, to investigate the impact on combustion behavior.
Table 7: Principal results from the reburning, advanced reburning and combustion without circulating mass experiments.
Parameters Unit Reburning Advanced
reburning Comb. without circ. mass
λ [-] 1.11 - 1.21 1.12 - 1.26 1.52 - 1.82
Alt. fuel % energy 49 - 64 36 - 56 28 - 57
q kW 37.5 - 53.1 35.7 - 55.8 20 - 29
τg sec. 0.64 - 0.69 0.63 - 0.67 0.86 - 1.06
vriser m/s 6.3 - 6.8 5.9 - 6.8 4.0 - 5.0
Re [-] 5300 - 5700 5570 - 5670 4520 - 4810
εriser [-] 0.99996 0.99996 -
Gs kg/h 29 29 -
NOsupplied mg/Nm³ 1931 72 - 1457 74 - 1450
Calsination % 70 - 86 83 - 87 -
SO3 wt% 0.36 - 0.6 0.075 - 0.45 -
Cl wt% 0.01 - 0.14 0.01 - 0.01 -
Results and discussion of the laboratory experiments
The principal laboratory experiments showed:
• almost no connection between the level of NO supplied and the CO and TOC emissions,
• that larger particle size gave increased CO emissions,
• that increasing the particle size changes the ranking of fuels with highest CO emittance and
• increasing CO and TOC emissions during reburning with animal meal in contrast to combustion in a lean environment (without circulating mass).
Combustion of animal meal showed increased CO and TOC emissions similar to the full-scale experiments. However, when animal meal is burnt in a lean environment the CO and TOC emissions are low and the competition between Equation 1 and 2 seems to be of minor importance.
A probable cause is that:
• for lean flames, the critical amine free radical NH2 is reduced directly or through the NNH intermediate, and/or
• the increased amount of hydrogen radicals increases the production of OH radicals.
6 The experiments were executed with circulating mass that had a mixture of 50/50 raw meal and Millisil®M6. The loading of circulating mass was about 67 % of the theoretical input in comparison to the precalciner /7/.
Reburning
The reburning experiment was executed with the following fuels and mixtures:
• refuse derived fuel (RDF),
• animal meal (AM),
• coal,
• RDF/AM (75/25 wt %),
• coal/AM (75/25 wt %), and
• CC/AM (25/75 wt %).
The experiments were executed with about 1930 mg /Nm³ of NO supplied into the riser inlet. The development between the reburning fuels is shown in Figure 6.
The reburning experiments showed, at conditions similar to a precalciner, the following order of optimal NO reducing fuels and mixtures:
• RDF/AM ( up to 41.4 % improvement related to coal),
• RDF ( up to 39.5 % improvement related to coal),
• CC/AM (up to 34.5 % improvement related to coal),
• AM (up to 25 % improvement related to coal),
• coal/AM (up to 18.3 % improvement related to coal), and
• coal.
Figure 6: Main plot: NOsupplied with respect to NOout/NOsupplied ratio during reburning experiments with RDF, coal, AM, RDF/AM, coal/AM and CC/AM. Subplot: C.fix/Volat. ratio with respect to
NOout/NOsupplied ratio.
Advanced reburning
The advanced reburning rich experiment was executed with:
• 25 wt % urea and RDF (RDF25),
• 5 wt % urea and RDF (RDF5), and
• 5 wt % urea and coal (coal5).
The experiments were executed with a four-step variation in the NO supplied (72 - 1457 mg /Nm³) with the development shown in Figure 7. Figure 8 show the urea/NOsupplied ratio with respect to NOout. The advanced reburning experiments showed that:
• the surplus of urea, which did not contribute to a reduction in the NO and emitted as ammonia, is found as a increase in the NO2 and N2O,
• RDF is a better fuel for advanced reburning compared to coal,
• increased amount of NOsupplied improves (lower) the ratio NOout / NOsupplied and
• the NO reduction depends on the ratio of (urea/ NOsupplied), with an optimum NOx reduction larger than 5, see Figure 8.
The advanced reburning showed, at conditions similar to a precalciner (1000 mg/Nm³ NOsupplied), the following order of optimal NO reducing fuels and urea mixtures:
• RDF5 (up to 60 %),
• RDF25 (up to 50 %), and
• coal5( up to 8 % )
Figure 7: Main plot: NOsupplied with respect to NOout/NOsupplied ratio during advanced reburning experiments with 25 wt % urea and RDF; 5 wt % urea and RDF; 5 wt % urea and coal. Subplot:
Close-up of the main plot.
Figure 8: The urea/NOsupplied ratio with respect to NOout without consideration of types of alternative fuel.
Combustion without circulating mass
The experiments without circulating mass was executed with:
• RDF,
• coal,
• AM, and
• SHW.
The experiments were accomplished with a four-step variation of the NO supplied (74 - 1450 mg /Nm³). There was no supply of tertiary air during the experiments without circulating mass. The overall excess air ratio was 1.52 to 1.82. Figure 9 and 10 show the development of the experiment without circulating mass.
Figure 9: Main plot: NOsupplied with respect to NOout/NOsupplied ratio during experiments without circulating mass. Experiments with RDFd<1mm, Coad<1mm, AMd<1mm, SHWd<2mm, Coald<2mm, AMd<2mm. Subplot: Close-up of the main plot.
Figure 10: NOsupplied/NOout ratio with respect to NOout during experiments without circulating mass.
Experiments with RDFd<1mm, Coald<1mm, AMd<1mm, SHWd<2mm, Coald<2mm, AMd<2mm.
Two sets of particle sizes were tested to investigate the impact on combustion behaviour. The particles was ground down into two groups:
• dp < 1, and
• dp < 2 mm
The combustion without circulating mass experiments showed that:
• smaller particles gave larger NO reduction - almost the double at the governing conditions,
• heterogeneous reactions are possibly the main reason for the NO reduction with increasing NOsupplied during combustion of coal, RDF and SHW. The increased N2O emissions during coal combustion are probably due to the heterogeneous reactions. However, a combination with homogenous reactions are likely,
• homogenous reaction is possibly the main reason for the NO reduction with increasing NOsupplied during combustion of animal meal. Hence the increased N2O emissions for animal meal are probably due to the homogenous reaction. Animal meal is, as mentioned above, probably governed by the ammonia released during volatilization, but a combination with heterogeneous reduction reactions are likely,
• the different behavior of animal meal is due to the assumed release of ammonia during volatilization, and
• increased amount of NOsupplied improves (lower) the ratio NOout/NOsupplied.
The combustion without circulating mass showed, at conditions similar to a precalciner (1000 mg/Nm³ NOsupplied), the following order of optimal NO reducing fuels:
• AMd<1 (up to 45 %),
• coald<1 (up to 18 %) and
• RDFd<1 (up to 6 %).
Following, AMd<2, SHWd<2 and coald<2 break even (NOout/NOsupplied=1) at similar conditions.
Flow calculations
Euler-Euler granular multiphase simulations are accomplished with FLUENT 4.0 in order to verify the assumption of dilute flow and the importance of particle-particle collisions in a precalciner.
Euler-Langrangian simulations of the precalciner are performed as well with FLUENT 6.0. The particle-trajectory, retention time and species concentration are calculated successfully. The simulations reveal important information for further utilization of alternative fuel in the precalciner.
A structured grid are generated in GAMBIT with the number of about 240,000 cells, see Figure 11.
The model height is 42.8 meter with a width at the section of inlets, and diameter at the mid- and top region of 3.56×3.27, 3.74 and 2.61 meter, respectively.
Figure 11: The figure shows the precalciner model with its inlets and outlets. The alternative fuel and raw meal inlet on the tertiary gas side are located opposite of the corresponding inlets and is not shown in the figure.
The Euler-Euler granular multiphase simulations showed that the particles:
• volume fraction is below 1⋅10-3, which is in accordance to previous characteristics,
• rise in two plumes from the injection points until it disperse between the first and second bend and increase the particles volume fraction close to the wall,
• obstructs part of the tertiary- and kiln gas inlet, which increases the local gas velocity, before it disperse and become uniformly distributed in the upper parts and
• that the kinetic energy turbulence can be seen to follow the raw meal volume fraction. Hence, the particles seem to enhance turbulence.
The Euler-Langrangian simulations showed:
• delayed volatilization on the kiln gas side, which most likely can be due to the suppressing effect of raw meal in the up stream giving a lower heat transfer. Replacement of the raw meal inlet could improve the volatilization, since the raw meal inlet is at the similar level as the alternative fuel inlet, which is in contrary to the tertiary kiln gas side,
• low degree of mixing between the kiln- and the tertiary air gas side, which represents poor oxidizing conditions for the alternative fuel on the kiln gas with a delayed burn up, see Figure 12. However, it is necessary with a low degree of mixing to obtain a successful utilization of reburning and advanced reburning in the precalciner. A reduction (optimization) of the reburning zone will improve the delayed burnout,
• that the ratio of solids- to gas retention time τs/τg is about 2 in the precalciner and half the size compared to previous experiments in a FLS-ILC calciner /6/. A comparison between previous work on CFBC reactors and precalciners with these simulations, show that particle retention time has to be considered for the specific precalciner. The back mixing / core-annulus flow, is low in this precalciner due to the high gas velocity.
Figure 12: Contours of mass fraction of O2 during Euler-Langrangian simulations.
Conclusion
It has been designed, constructed and executed experiments in a CFBC reactor accompanied with full scale experiments in the precalciner at kiln 6 in Brevik. Furthermore, it has been executed CFD simulations to verify the precalciner regime and to investigate the flow patterns influence on the burn up. Investigations on the following fuels, and blends of them, has been executed:
• Refuse derived fuel,
• Animal meal,
• Solid hazardous waste,
• Char coal, and
• Coal.
This work showed that:
• Laboratory experiments, in a similar regime as in a precalciner, demonstrate to be well suited for characterization of alternative fuels in a precalciner regime.
• Experiments with and without circulating mass in the laboratory CFBC reactor justify the importance of executing combustion experiments in similar environment to a precalciner.
• Advanced reburning and reburning are well suited for NOx reduction in a precalciner during utilization of SHW and animal meal. RDF shows the best capacity to reduce NOx during reburning and advanced reburning in laboratory experiments. However, previous full-scale experiments show larger NOx reduction for animal meal compared to RDF.
• Animal meal is believed to follow the reduction route of SNCR or advanced reburning.
• Increasing amount of NOsupplied at the reactor inlet improves the ratio of NOout / NOsupplied for all fuels during laboratory experiments with advanced reburning and combustion without circulating mass.
• Increasing ratio of urea/ NOsupplied lower the NOx emissions. A conclusive optimum is not possible to make without further investigations, but an assumption shows an optimum between 8 - 10 and 2.6 - 3.0 in the laboratory experiments and the full-scale experiments, respectively.
• Increased CO emissions during advanced reburning and reburning (with animal meal) is most likely to be due to the competition for the OH radical during oxidation of CO and reduction of NH3. However, combustion of animal meal, in a lean environment without circulating mass, showed low CO emissions.
• Heterogeneous reactions reduce NO with increasing amount of NOsupplied. This was shown during combustion without circulating mass in a lean environment in the laboratory
experiments. This finding emphasize that NO reduction is governed by several phenomenon.
• Smaller particles reduce CO and NO emissions.
• CFD simulations reveal important information about the flow regime and its importance to improve burn up.
References
[1] Xu, H., Smoot, L. D. & Hill, S. C. (1998), ‘A kinetic model for NOx reduction by advanced reburning’, Energy & Fuels 12, 1278-1289.
[2] Hämäläinen, J. (1995) ‘Effect of Fuel Composition on the Conversion of Fuel-N to Nitrogen Oxides in the Combustion of Small Single Particles’, PhD thesis, Department of Chemistry, University of Jyväskylä.
[3] Zevenhoven, R., Axelsen, E. P., Kilpinen, P., Hupa, M., Elomaa, M. & Liukkonen, V.-P.
(2000), ’The behaviour of nitrogen from polymers and plastics in waste- derived fuels during combustion’, Technical Report Final report, Åbo Akademi University, Process Chemistry Group.
[4] Desroches-Ducarne, E., Dolignier, J. C., Marty, E., Martin, G. & Delfosse, L. (1998),
‘Modelling of gaseous pollutants emissions in circulating fluidized bed combustion of municipal refuce’, Fuel 77, 1399-1410.
[5] Axelsen, E. P. (2001), ’Laboratory circulating fluidizing bed combustor – design and
preliminary experiments – comparative study to the precalciner at Norcem Brevik’, Technical Report, Scancem Int. ANS.
[6] Hundebøl, S. & Kumar, S. (1987), ‘Retention time of particles in calciners of the cement industry’, Zement – Kalk – Gips, 422-425.
